The SEC brought a civil action against Scott Ginsburg alleging violations of § 10(b) and § 14(e) of the Securities and Exchange Act, 15 U.S.C. §§ 78j(b), 78n(e), and accompanying Rules 10b-5 and 14e-3, 17 C.F.R. §§ 240.10b-5, 240.14e-3. The allegations were that Ginsburg had communicated material nonpublic information to his brother Mark Ginsburg and to his father Jordan Ginsburg regarding EZ Communications, Inc., and Katz Media Group, and that Mark and Jordan had traded on EZ stock using that information.1

The case was tried to a jury which found that Ginsburg had violated the insider trading provisions. The district court initially ordered him to pay $1,000,000 in penalties but denied the SEC's request to enjoin him from violating securities laws in the future. Later the district court granted Ginsburg's renewed motion for judgment as a matter of law and vacated the judgment against him, because it concluded that the evidence was insufficient to permit a reasonable jury to find that he had tipped off his brother or father about inside information.

The SEC appeals from that judgment as a matter of law, and also contends that if we reverse it, we should direct the district court to enjoin Ginsburg from violating securities laws in the future. In addition to defending the district court's judgment on the specific insufficiency of evidence ground given for it, Ginsburg offers two other grounds for affirming it. First, he contends that even if the evidence is sufficient to show he communicated insider information to Mark or Jordan, it is insufficient to show the information's materiality. Second, he contends the evidence is insufficient to show substantial steps were taken toward a tender offer of Katz, and Ginsburg's knowledge of the tender offer. Finding merit in the SEC's position and none in Ginsburg's, we conclude that the judgment is due to be reversed, the civil penalty reinstated, and an injunction entered against future violations of the securities laws by Ginsburg.

The nature of a judgment as a matter of law and our review of it is such that we take the evidence at trial in the light most favorable to the party who won before the jury only to have its victory taken away by the court. Russell v. North Broward Hosp., 346 F.3d 1335, 1343 (11th Cir.2003). We draw from the evidence all reasonable inferences in support of the verdict, because the jury could have done so. United States v. Gregory, 730 F.2d 692, 700 (11th Cir.1984).

Ginsburg was chairman and CEO of Evergreen Media Corporation, which owned and operated a number of radio stations. In 1996 Evergreen became interested in acquiring EZ, a corporation that owned radio stations. Ginsburg met with EZ's CEO Alan Box on Friday, July 12, 1996, and Box told him EZ was considering several "strategic alternatives." Before the meeting no decision had been made to sell the company. On Sunday evening, July 14, Ginsburg called Mark at 10:02 p.m., and they spoke for 26 minutes. The next day, Monday, July 15, Mark bought 3800 shares of EZ stock. Mark spoke with Jordan in person and on the telephone over the next few days, and they admit they discussed the purchase of EZ stock.

On July 15, EZ's investment banker called Ginsburg and they discussed the possibility of Evergreen submitting a bid on EZ. Ginsburg asked to be sent a confidentiality agreement that he could sign in order to receive financial information about EZ. He received the confidentiality agreement on July 16. That same day Jordan purchased 20,000 shares of EZ. After signing the confidentiality agreement, Ginsburg received the financial information about EZ on July 17. On July 18, a four minute telephone call was placed from Ginsburg's office to Mark's home.

On July 24, EZ's investment banker faxed a letter to Ginsburg stating that EZ bids were due by July 26. At 7:40 a.m. July 25, a call was placed from Ginsburg's cell phone to his parents' home. Later that day Mark purchased 3200 shares of EZ stock for Mark and his wife's joint account. On July 26, Mark purchased 4300 shares of EZ stock for the same joint account and 7500 shares for a trust account for his son. Also on July 26, Ginsburg, on Evergreen's behalf, submitted a written offer to acquire EZ for cash and stock. On Sunday, July 28, a ten minute call was placed from Ginsburg's home to Mark's home. On Monday, July 29, Jordan bought 5000 shares of EZ stock, and an hour later Mark bought 30,000 shares.

Evergreen's bid for EZ ultimately fell through, but on August 5, 1996, EZ announced its merger with another radio company, at which time the price of EZ stock rose 30%. Between July 15 and July 29, Mark had bought 48,800 shares of EZ and Jordan had bought 25,000 shares, which increased in value $664,024 and $412,875 respectively by August 5, 1996.

In early 1997, Ginsburg's company, Evergreen, was in the process of merging with Chancellor Broadcasting. On March 20, 1997, Ginsburg attended a meeting with senior executives of Katz Media Group and Hicks, Muse, Tate, & Furst, an investment firm that owned a majority interest in Chancellor Broadcasting, at which a possible acquisition of Katz by Chancellor was discussed. Tom Hicks, Chairman and CEO of Hicks, Muse, appointed a due diligence team headed by Ginsburg. A confidentiality agreement was executed April 7, 1997, and due diligence began. On June 16, 1997, Stuart Olds, a Katz executive, met with Ginsburg. Olds encouraged Ginsburg to call Katz chairman Tom Dean to discuss the purchase of Katz. Olds also told Ginsburg that Katz was having discussions with other companies and Ginsburg would have to act quickly.

That same evening, June 16, a call was placed from a cell phone registered to Ginsburg to a phone registered to Mark. The next day, June 17, Mark bought 150,000 shares of Katz. On July 14, Evergreen/Chancellor announced it would acquire Katz through a tender offer for Katz stock at $11 a share. On July 16 or 17, 1997, Mark sold 132,500 shares of Katz and tendered the rest, resulting in a total profit of $729,000.

We review a decision to grant a motion for judgment as a matter of law de novo, applying the same standards used by the district court. SEC v. Adler, 137 F.3d 1325, 1340 (11th Cir.1998). A judgment as a matter of law is warranted only "[i]f during a trial by jury a party has been fully heard on an issue and there is no legally sufficient evidentiary basis for a reasonable jury to find for that party on that issue." Fed.R.Civ.P. 50(a)(1). That means, as we have already said, that we review the evidence, and the inferences arising therefrom, in the light most favorable to the non-moving party. We "may not weigh the evidence or decide the credibility of witnesses." Adler, 137 F.3d at 1340. However, the nonmoving party "must provide more than a mere scintilla of evidence to survive a motion for judgment as a matter of law." Isenbergh v. Knight-Ridder Newspaper Sales, Inc., 97 F.3d 436, 439 (11th Cir.1996).

We review denial of equitable relief for abuse of discretion. Preferred Sites, LLC v. Troup County, 296 F.3d 1210, 1220 (11th Cir.2002). Any factual findings made by the district court with regard to its denial of an injunction are reviewed for clear error. Id.

In order to establish liability under § 10(b) and § 14(e) of the Securities and Exchange Act and accompanying Rules 10b-5 and 14e-3, the SEC must prove that Ginsburg acted with scienter, "`a mental state embracing intent to deceive, manipulate, or defraud.'" SEC v. Adler, 137 F.3d 1325, 1340 (11th Cir.1998) (quoting Aaron v. SEC, 446 U.S. 680, 695-96, 100 S. Ct. 1945, 1955, 64 L. Ed. 2d 611 (1980)). Scienter requires that the insider (or tippee, if the trader is not the insider) possess material nonpublic information at the time of the trade. Id. at 1340. In addition, it requires that the material non-public information be used in a trade. Id. Proof of knowledge of such information at the time of a trade "gives rise to a strong inference of use." Id.

The SEC must prove violations of § 10(b) and § 14(e), and their supplementary Rules, by a preponderance of the evidence, and may use direct or circumstantial evidence to do so. Herman & MacLean v. Huddleston, 459 U.S. 375, 390 & n. 30, 103 S. Ct. 683, 691-92 & n. 30, 74 L. Ed. 2d 548 (1983). "Circumstantial evidence has no less weight than direct evidence as long as it reasonably establishes that fact rather than anything else." Burrell v. Bd. of Trustees of Ga. Military College, 970 F.2d 785, 788 (11th Cir.1992).

The parties disagree about which body of precedent controls the sufficiency of the evidence issue upon which the district court granted judgment as a matter of law. The sufficiency issue in general involves the circumstances in which it will be inferred from A's act following a conversation with B, who knew a given fact, that A had been informed of that fact when he acted. As it arises in insider trading cases, the more specific issue is when it may be inferred from a trade in stock by A, following a conversation with insider B, that B disclosed inside information to A who acted upon it. The SEC argues, logically, that the sufficiency of the evidence and the permissibility of inferences that may be drawn from the evidence in this insider trading case are governed by our insider trading decisions, especially SEC v. Adler, 137 F.3d 1325 (11th Cir.1998), which is the closest of those cases to the facts we have here.

Ginsburg argues, and the district court concluded, that four employment retaliation decisions, which he views as inconsistent with Adler on the issue of what may be inferred from conversations of an actor with one in the know, control instead. See Burrell, 970 F.2d 785 (11th Cir.1992); Clover v. Total Sys. Servs., Inc., 176 F.3d 1346 (11th Cir.1999); Brungart v. BellSouth Telecomm., Inc., 231 F.3d 791 (11th Cir.2000); Brochu v. City of Riviera Beach, 304 F.3d 1144 (11th Cir.2002). The last three of those employment decisions cannot possibly trump Adler to the extent of any inconsistency. Where prior panel decisions conflict, we follow the first one released, Cohen v. Office Depot, Inc., 204 F.3d 1069, 1072 (11th Cir.2000) ("where two prior panel decisions conflict we are bound to follow the oldest one"); United States v. Dailey, 24 F.3d 1323, 1327 (11th Cir.1994) (where there is an intra-circuit conflict, "the earliest panel opinion resolving the issue in question binds this circuit until the court resolves the issue en banc"), and Adler came out before any of those three decisions.

However, Burrell, one of the four employment decisions that Ginsburg cites, and upon which the district court relied to award him judgment as a matter of law, preceded Adler. If an evaluation of the sufficiency of the evidence should not take into account the specific factual context from which that evidence arose, then we would be bound to follow Burrell and disregard Adler to the extent of any inconsistency. But it should, so we are not.

This is what we are talking about. Decision maker A comes into contact with information possessor B and soon thereafter engages in conduct C. When will that factual scenario support an inference that, despite their denials, A was told the information by B and on that basis A did C? More to the point, does an earlier decision of this Court, concluding that scenario would not support an inference the information was communicated and acted upon in an employment retaliation case, compel us to conclude that the same scenario would not support an inference of communicated information and action based upon it in insider trading cases?

We think the earlier employment retaliation decision in Burrell does not control this case, because insider trading cases are different from employment retaliation cases. The context in which the facts arise and the strength of the competing inferences can differ. As a result, evidence that may appear to be materially identical for purposes of determining whether a decision maker knew a particular fact can actually have different probative force in an insider trading case than in an employment retaliation case.

There are many sound, non-retaliatory business reasons to take some job action that is challenged as retaliatory. That multitude of potential reasons dilutes the strength of any inference that because a decision maker took an action against an employee he must have been told of a fact which could have led him to take the action for a prohibited reason.

By contrast, people do not make large stock trades for as many reasons as businesses take job actions. Although there are exceptions, people generally buy when they believe the price of a stock is going up and sell when they believe it is going down (either absolutely or relative to the expected performance of other stock). The factfinder in an insider trading case need only infer the most likely source of that belief. The temporal proximity of a phone conversation between the trader and one with insider knowledge provides a reasonable basis for inferring that the basis of the trader's belief was the inside information. The larger and more profitable the trades, and the closer in time the trader's exposure to the insider, the stronger the inference that the trader was acting on the basis of inside information. The magnitude of the incentive to trade on insider information is illustrated by the trades that were made in this case. In less than a month Jordan made $412,875 by trading EZ stock in the direction someone with knowledge of the insider information his son possessed would have, and Mark made a total of $1,393,022 by trading EZ and Katz stock as someone privy to the insider information of his brother would have.

It is not at all clear that the same considerations apply with equal force in job discrimination cases. The inference that a job action was based on a retaliatory motive which arose from imparted information may well be weaker than the comparable inferences in insider trading cases for several reasons. For one thing, the incentive to tip and to act on tipped information is usually a great deal stronger than the incentive to impart and act upon information about an employee engaging in legally protected conduct. We expect that most people would rather make $412,875 or $1,393,022 in a short period trading stocks than they would like to see an employee be punished for something the employee had a legal right to do.

Because it is far from clear that employment retaliation cases are interchangeable with insider trading cases, the district court should have looked to the more specifically applicable precedent instead of regarding it as wrongly decided in light of decisions that had nothing to do with insider trading. Cf. Rodriguez de Quijas v. Shearson/Am. Express, Inc., 490 U.S. 477, 484, 109 S. Ct. 1917, 1921-22, 104 L. Ed. 2d 526 (1989) ("If a precedent of this Court has direct application in a case, yet appears to rest on reasons rejected in some other line of decisions, the Court of Appeals should follow the case which directly controls, leaving to this Court the prerogative of overruling its own decisions."). It is insider trading decisions, instead of employment retaliation cases, to which we will look for guidance in deciding this case.

In the Adler insider trading case, Richard Adler, an outside director of Comptronix Corporation, attended a November 15 board meeting at which he was asked to investigate the activities of two Comptronix executives for potential fraud. Adler, 137 F.3d at 1329. The next day, November 16, Adler's friend and associate Harvey Pegram called Adler's home at 7:53 a.m. The call lasted 72 seconds. At 7:55 a.m. Pegram called his wife at home. At 8:07 a.m. Pegram's wife called their stockbroker and placed an order to sell 50,000 shares of Comptronix, and over an eight-day period beginning with that telephone call they sold a total of 150,000 shares. On the evening of November 16 at 8:02 p.m., Pegram called his business associate Philip Choy in Hong Kong. At 9:39 p.m. Choy faxed his stockbroker an order to sell 5000 shares of Comptronix. Another associate of Pegram, Domer Ishler, called Adler on November 15 and again on November 23. On November 24, Ishler purchased 300 put options on Comptronix stock. On November 25 Comptronix announced that certain officers had been suspended for overstating gross profits, and Comptronix common stock lost 72% of its value. Id. at 1329. The Pegrams avoided losses of $2,315,375, Choy avoided losses of $75,000, and Ishler realized gains of $368,750. Id. at 1331.

The SEC brought a civil action against Adler, Pegram, Choy, and Ishler for violations of § 10(b) and Rule 10b-5. Id. at 1327. After a hung jury, the district court granted judgment as a matter of law to Pegram, Adler and, Choy, and summary judgment to Ishler. Id. at 1331-32. The court concluded that the phone calls raised a "possible reasonable inference" that Pegram received inside information from Adler, but any such inference was rebutted by Pegram's evidence of a preexisting plan to sell his Comptronix stock. The circumstances related to the other defendants were similar, and the district court made the same determination for each one. Id. at 1331-32.

We reversed, concluding that the phone calls and stock sales raised a reasonable inference that Pegram was tipped by Adler, and thus that Pegram possessed nonpublic information. We noted that a credible and innocent explanation for the timing of calls and sales can rebut the inference of possession of information. Id. at 1341. Nevertheless, though Pegram provided "strong evidence" that his plan to sell Comptronix stock predated the alleged tip, and evidence of innocent explanations for the relevant telephone calls, we stated that the jury is not required to believe the defendant's explanations. "[T]hese fact-intensive issues should be decided by a jury, which is in the position to observe the demeanor of witnesses and make appropriate credibility determinations." Id. at 1342. Choy's calls and sales similarly created a reasonable inference, as did Ishler's. Id. at 1342-43.

In Adler the calls/trades pattern repeated twice on one day and once again the next week. In this case there is evidence of one clear call/trade pattern concerning EZ stock (the July 14 call from Ginsburg to Mark followed the next day by his purchase of 3,800 shares), and one concerning Katz stock (the June 16 call from Ginsburg to Mark followed the next day by his purchase of 150,000 shares). The other EZ calls match less well with trades. The July 25 call to Jordan was followed by a purchase by Mark, and the July 28 call to Mark followed with a purchase by Jordan. But because Mark and Jordan admitted discussing EZ throughout that period, the mismatch of calls and trades is not a big problem. The multiple occurrences of the pattern in this case are similar enough to those in Adler. This case does have stronger evidence suggesting that conversations between the parties were more frequent,2 and that the types of trades made were not unique,3 and it also features longer spans of time between the conversations and trades. Ginsburg offered evidence of public information about the companies as motivation for Mark's trading,4 argued that the trades were consistent with prior trading history, and put forward innocent explanations for the calls. The district court commented that "it is plausible that the investments... were driven not by tips but rather by public knowledge." It is also plausible that they were driven by insider information. And it was up to the jury to choose between those competing plausible theories of fact.

The jury was free to disbelieve Ginsburg's evidence just as the Adler jury was free to disbelieve what we characterized as the "strong" evidence of a preexisting stock trade plan in Adler. Evidence of the innocent explanations for the calls between the parties in this case and of Mark's trading habits is not enough to justify overturning the jury's verdict. If it were otherwise, family members who regularly traded in a particular stock or type of stock could trade based on insider information with impunity.

The district court stated that "[t]he phone records are insufficient to compel an inference that Scott Ginsburg conveyed material, non-public information to Mark," but that is not the issue. The SEC did not have the burden of putting in evidence that compelled the inference Ginsburg conveyed nonpublic information to Mark. All it was required to do was put in evidence that reasonably permitted that inference. It did that. The call/trade pattern occurrences coupled with the jury's right to disbelieve the innocent explanations of the calls and trades are enough to support the verdict.

Ginsburg argues that even if sufficient evidence was presented that he tipped Mark and Jordan, the district court's decision to grant judgment as a matter of law can be affirmed on two independent grounds. Though not reached by the district court, these arguments were presented in Ginsburg's renewed Rule 50 motion, and they raise legal issues over which we would exercise de novo review, so we may decide them in the first instance.

Ginsburg contends that the SEC did not provide sufficient evidence to permit a reasonable jury to find that the information tipped was material and nonpublic, as required by Rule 10b-5. "An omitted fact is material if there is a substantial likelihood a reasonable shareholder would consider it important in deciding how to vote." TSC Indus., Inc. v. Northway, Inc., 426 U.S. 438, 449, 96 S. Ct. 2126, 2132, 48 L. Ed. 2d 757 (1976). Materiality is proved by showing a "substantial likelihood that the disclosure of the omitted fact would have been viewed by the reasonable investor as having significantly altered the `total mix' of information made available." Id.

The insider information in this case meets the materiality standard. In Basic Inc. v. Levinson, 485 U.S. 224, 238, 108 S. Ct. 978, 987, 99 L. Ed. 2d 194 (1988), the Supreme Court held that preliminary merger talks can be material well before any agreement is reached. The materiality of an uncertain prospective event "depend[s] at any given time upon a balancing of both the indicated probability that the event will occur and the anticipated magnitude of the event in light of the totality of the company activity." Id. The Court explained that "a factfinder will need to look to indicia of interest in the transaction at the highest corporate levels," and consider factors such as "board resolutions, instructions to investment bankers, and actual negotiations between principals or their intermediaries." Id. at 239, 108 S. Ct. at 987. The determination of materiality "requires delicate assessments of the inferences a `reasonable shareholder' would draw from a given set of facts and the significance of those inferences to him, and these assessments are peculiarly ones for the trier of fact." TSC Indus., Inc., 426 U.S. at 450, 96 S. Ct. at 2133. "In an analogous context, the jury's unique competence in applying the `reasonable man' standard is thought ordinarily to preclude summary judgment in negligence cases." Id. at 450 n. 12, 96 S. Ct. at 2133 n. 12.

Alan Box, EZ's CEO, and Ginsburg discussed a specific share price, and Ginsburg knew both that Evergreen was interested in acquiring EZ and that EZ was talking to other companies. The jury was free to find that a reasonable investor would view Evergreen executives talking to EZ executives about a possible acquisition at the specific price of $50 a share as altering the total mix of information available. A merger is an event of considerable magnitude to an investor, and preliminary merger negotiations constitute concrete steps indicating an increasing possibility of a merger occurring. As for the nonpublic nature of the evidence, the jury heard testimony from Ron Peale, EZ's CFO, that the talks were confidential. The jury could recognize as material the difference between public information that EZ was having conversations about a possible sale and about potential sale prices, see supra n. 4, and nonpublic information about a private meeting between executives and the specific share price they discussed confidentially.

As to Katz, Ginsburg was clearly privy to information gleaned under a confidentiality agreement. The jury had before it sufficient evidence to conclude that Ginsburg's June 16 meeting with Olds indicated to Ginsburg that a deal had to happen fast. The jury could reasonably infer that the information was material and nonpublic.

Ginsburg also argues that there was not enough evidence to permit an inference that, at the time of his alleged Katz tip on June 16: (1) substantial steps had been taken to commence a tender offer, as required by Rule 14e-3; and (2) Ginsburg knew of the tender offer, as Ginsburg claims is required by Rule 14e-3. Rule 14e-3 prohibits purchase or sale of a company's stock by "any person who is in possession of material information relating to [a] tender offer" if an "offering person" has taken "a substantial step or steps to commence, or has commenced, a tender offer." 17 C.F.R. § 240.14e-3(a). Rule 14e-3(d) prevents any insider of the target or acquiring company, or any person in possession of material information relating to a tender offer, from communicating that information to another likely to trade on it, after substantial steps have been taken. The SEC release accompanying Rule 14e-3 states:

[S]ubstantial step or steps to commence a tender offer include, but are not limited to, voting on a resolution by the offering person's board of directors relating to the tender offer; the formulation of a plan or proposal to make a tender offer by the offering person or the person(s) acting on behalf of the offering person; or activities which substantially facilitate the tender offer such as: arranging financing for a tender offer; preparing or directing or authorizing the preparation of tender offer materials; or authorizing negotiations, negotiating or entering into agreements with any person to act as a dealer manager, soliciting dealer, forwarding agent or depository in connection with the tender offer.

Tender Offers, Exchange Act Release No. 17,120, 20 SEC Docket 1350 n. 33 (Sept. 4, 1980).

In this case, Katz CEO Tom Olson had sent a letter to Tom Hicks, CEO of the majority shareholder of Chancellor, on February 24, 1997, asking whether Chancellor was interested in acquiring Katz. On March 20, Olson, Hicks, Olds, Ginsburg, and others met to discuss the possible acquisition. Hicks was interested in Chancellor acquiring Katz. He appointed a due diligence team, and a confidentiality agreement was signed on April 7. Ginsburg met with Olds on June 16, and Olds told him that Katz was in discussions with other companies and Chancellor needed to act quickly. Testimony at trial indicated that the parties had not settled on a tender offer as the form of the transaction until the last few days before the deal was announced on July 14.

These activities do not fall into the specifically enumerated examples of activities described as "substantial steps" in the SEC release. However, the release makes it clear that the examples listed are only that; they are not a complete list of "substantial steps." Neither this Court nor the Supreme Court has defined "substantial steps," but other circuits have confronted the issue. In SEC v. Maio, 51 F.3d 623 (7th Cir.1995), the Seventh Circuit concluded that a meeting between officers of the target and acquiring companies, held after the target had solicited an offer from the acquiring company, which was "much more serious than any previous discussion between the parties," and which was followed the next day by the onset of the due diligence process, constituted "substantial steps." Id. at 636. A confidentiality agreement had not yet been entered at the time of the alleged tip in that case. Id. In SEC v. Mayhew, 121 F.3d 44 (2d Cir. 1997), where the merging companies had retained a consulting firm, signed confidentiality agreements, and held meetings between top officials, the Second Circuit concluded that the "substantial steps" requirement was satisfied, despite the fact that the companies had not settled on a tender offer as the form of the merger. Id. at 53.

In this case there was a meeting between executives, which was followed by due diligence procedures, a confidentiality agreement, and by a meeting between Ginsburg and Olds — from which Ginsburg realized that the deal had to go down fast. These activities, which did result in a tender offer, were substantial steps for purposes of Rule 14e-3. Were it otherwise, liability could be avoided by taking care to tip only before the formal steps finalizing the acquisition are completed, leaving a substantial gap between the acquisition of inside information and the regulation of its disbursement.

Rule 14e-3, by its terms, does not require that the offender know or have reason to know that the information relates to a tender offer, so long as the information in fact does relate to a tender offer and the offender knows or has reason to know the information is nonpublic and was acquired by a person with the required status. The accompanying SEC release explicitly states that there is no "knows or has reason to know" standard attached to the Rule's requirement that the information relate to a tender offer. Tender Offers, Exchange Act Release No. 17,120, 20 SEC Docket 1350 (Sept. 4, 1980). We agree with the First Circuit that we should defer to the SEC's interpretation of its Rule. See SEC v. Sargent, 229 F.3d 68, 78-79 (1st Cir.2000). Under United States v. O'Hagan, 521 U.S. 642, 673, 117 S. Ct. 2199, 2217, 138 L. Ed. 2d 724 (1997), we owe the judgment of the SEC expressed in its release "more than mere deference or weight." It follows that Ginsburg's substantial steps argument fails.

Having decided that we should reverse the judgment and reinstate the jury verdict, we come now to the SEC's contention that the district court abused its discretion in refusing to permanently enjoin Ginsburg from future violations of the Exchange Act. Our review is for an abuse of discretion, which occurs when "the district court has made a clear error of judgment or has applied an incorrect legal standard." Doe v. Chiles, 136 F.3d 709, 713 (11th Cir.1998).

In order to be entitled to injunctive relief the SEC had to show a reasonable likelihood that Ginsburg would violate the securities laws in the future. SEC v. Carriba Air, Inc., 681 F.2d 1318, 1322 (11th Cir.1982). The district court cited the proper factors to consider in making this determination: "the egregiousness of the defendant's actions, the isolated or recurrent nature of the infraction, the degree of scienter involved, the sincerity of the defendant's assurances against future violations, the defendant's recognition of the wrongful nature of his conduct, and the likelihood that the defendant's occupation will present opportunities for future violations." Id.; SEC v. Blatt, 583 F.2d 1325, 1334 n. 29 (5th Cir.1978).5

The district court applied the correct legal standard, so the question is whether it made a clear error of judgment in the result it reached applying that standard. We believe it did, because every one of the factors to be considered weighs in favor of enjoining Ginsburg. As CEO of Evergreen, he breached his fiduciary duty to his company and to its shareholders on two separate occasions for the financial gain of his family members. Deliberately tipping material nonpublic information for family members' financial gain is a bad thing, and doing it twice in a year is doubly so. See Blatt, 583 F.2d at 1334-35 (affirming injunction of a defendant whose conduct encompassed just the type of "knowing and intentional misconduct" § 10(b) was intended to proscribe, where defendant was the "mastermind" behind two violations). It is this type of intentional, knowing conduct, as opposed to more minor, technical violations, for which injunctions are reserved. See id. at 1335 (reversing injunction of defendant whose single violation was relatively minor); accord SEC v. Steadman, 967 F.2d 636, 648 (D.C.Cir.1992) (considering whether defendant's violation was "flagrant and deliberate or merely technical in nature").

Scienter is a required element of a § 10(b) violation, Adler, 137 F.3d at 1340, and thus the jury, in finding that such a violation occurred, found that Ginsburg acted with scienter. Though Ginsburg assured the district court he would not in the future "engage in any actions that will raise suspicion of illegal conduct," such promises to toe the line are "by no means dispositive." See Carriba Air, Inc., 681 F.2d at 1322 (affirming district court's injunction of defendant against future violation of securities laws despite defendant's assertions he would cease his wrongful conduct). And they are of very little worth at all where, as here, the defendant denies having done anything at all improper to begin with. Promising to stop doing wrong while denying any wrongdoing is the wrong way to establish that wrongdoing will not reoccur.

Ginsburg is currently chairman of the board of a public company, and in this position he has access to material nonpublic information, providing him with future opportunities to violate the securities laws. See Blatt, 583 F.2d at 1335 (affirming injunction of a defendant guilty of only one "reprehensible fraud" where his continuing investment opportunities strengthened the inference from his past conduct that he was likely to commit future violations).

In short, each Carriba factor cuts against Ginsburg, establishing a reasonable likelihood he will violate securities laws again. Cf. Carriba Air, Inc., 681 F.2d at 1322 (affirming district court's injunction of defendant where virtually all factors cut against defendant); Commodity Futures Trading Comm'n v. Sidoti, 178 F.3d 1132, 1137 (11th Cir.1999) (affirming district court's injunction against future violation of Commodity Exchange Act given likelihood of future violation). If no injunction is to be entered where every factor weighs in favor of an injunction, then the factors would be meaningless. They are not, and it was a clear error of judgment not to enjoin Ginsburg to refrain from violating the securities laws in the future.

The district court's grant of judgment as a matter of law is REVERSED and the case is REMANDED with instructions that the court reinstate the civil penalty of $1,000,000, and enjoin Scott Ginsburg from future violations of the securities laws and regulations.

I think it is useful from time to time to repeat something that Chief Justice Marshall wrote:

It is a maxim not to be disregarded, that general expressions, in every opinion, are to be taken in connection with the case in which those expressions are used.... The reason of this maxim is obvious. The question actually before the court is investigated with care, and considered in its full extent. Other principles which may serve to illustrate it, are considered in their relation to the case decided, but their possible bearing on all other cases is seldom completely investigated.

Cohens v. Virginia, 19 U.S. (6 Wheat.) 264, 399-400, 5 L. Ed. 257 (1821).

A case is no precedent for a proposition that was not in the mind of the court when the case was decided; given Burrell's facts, no reason exists to believe the court in Burrell had insider-trading in its mind when it decided Burrell. So, I agree that the district court erred in extending what it read as the holding in Burrell (an employment case) to a materially different kind of case (an insider-trading case) and, then, in concluding that in some way Burrell trumped our Adler decision which was an insider-trading precedent.

I concur in today's judgment, except for reversing the denial of the injunction. For the injunction, the district court applied the correct legal standard and had the important advantage over us of having seen the witnesses and parties firsthand as they testified and made representations to the district court. The grant or denial of injunctive relief "rests within the sound discretion of the trial court and will not be disturbed unless there has been a clear abuse of it." SEC v. Blatt, 583 F.2d 1325, 1334 (5th Cir.1978).* Especially considering that no finding was made that Scott Ginsburg was personally enriched as a result of the occurrences underlying this case and considering his assurance to the court that he would avoid similar conduct in the future, I cannot say that the district court abused its discretion in denying the injunction. Therefore, I would affirm the denial of the injunction (although if the district court had granted an injunction, I would likely have voted to affirm that decision).